Deep Learning-Based System for Detection of Lung Cancer Using Fusion of Features

The cancer detection is doing with the aid of the skilled expert docs and earlier tiers it may be helpful. The opportunity of human error must be there. Employing Deep Learning (DL) in the medical sector is very crucial due to the sensitivity of this field. This means that the low accuracy of the classification methods used for lung detection is a critical issue. This problem is accentuated when it comes to blurry medical images. Moreover, the low accuracy problem is accentuated when DL-based detection systems cannot manipulate the tumour from different angels of views. This paper presents the Adoptive Lung Cancer Detection (ALCD) system, which is built based on the Convolutional Neural Networks (CNN). The ALCD system uses an effective pre-processing phase, to ensure the quality of the medical images, depending on histogram equalization technique. In addition, the CNN is fed by features extracted using Scale Invariant Feature Transform (SIFT). Compared to the state-of-arte, the ALCD system shows better performance in terms of accuracy, sensitivity, and error rate.